Abstract: 

The rapid rise of Large Language Models is creating new industries, new categories of products, and forcing the tech ecosystem to rethink what is possible with these new systems. This talk explores a number of patterns for using LLMs as building blocks, beyond building shallow wrappers around commercial APIs or open source models. Learn how LLMs can be used to inject intelligence into software systems that do semantic search, text classification, retrieval-augmented generation, and more.

Bio: 

Jay Alammar, Through his popular machine learning blog, Jay has helped millions of engineers visually understand machine learning tools and concepts from the basic (ending up in NumPy, pandas docs) to the cutting-edge (The Illustrated Transformer, BERT, GPT-3).